Epidemiology/natural history

The prevalence of VHD in industrialised countries has been extrapolated from studies predominantly conducted in the USA,1 while the prevalence of rheumatic disease in sub-Saharan Africa is extrapolated from studies in North Africa. True figures need to be established nationally, while for rare causes of VHD (eg, carcinoid or antiphospholipid syndrome), this might be better done using international registries with standardised protocols. Serial echocardiography within these projects will improve our understanding of the contemporary natural history of VHD, which was previously determined in small cohorts of patients and generally with fewer comorbidities compared with the present.

Basic biology/physiology

The genetics and developmental biology of VHD are poorly understood. Collation of genetic analyses from established bio-banks and twin studies may identify new determinants of disease or its progression. Such techniques may also provide clues towards the development of treatments for challenging conditions such as endomyocardial fibrosis. Lipid-lowering therapy has not been successful in modifying the progression of calcific aortic stenosis (AS), although all large trials to date studied patients with relatively advanced disease. It remains possible that treating the disease earlier in its natural history will still be successful, but the roles of subsequent angiogenesis, inflammation and calcium metabolism require further elucidation to derive disease-modifying drugs. Recent work shows that there is coupling between the aortic valve and aorta which should be considered a single physiological unit.5 The implications of this on coronary flow, exercise capacity or the mechanism of symptom production are unknown but merit investigation.

Medical therapy

There is virtually no pharmacological therapy recommended in VHD other than vasodilators for chronic aortic regurgitation and even this has a conflicting evidence base.6 ,7 β-blockers may decrease regurgitant volume in mitral regurgitation8 (MR) and a randomised controlled trial (RCT) is needed to establish whether this can delay surgical intervention, particularly in high-risk patients. β-Blockers and angiotensin receptor blockers (ARBs) reduce events from aortic disease in Marfan syndrome, though there is no equivalent information for bicuspid valve disease. An observational study has suggested that ACE inhibitors or ARBs reduce the risk of events in AS.9 Controlled trials are now justified.

Timing of surgery in native VHD

Severe symptomatic VHD is fatal if untreated, but timely intervention can prolong survival. Recent advances in surgical repair and transcatheter techniques are tempered by uncertainties about when these should be used. There is a need for RCTs investigating whether immediate surgery is superior to ‘watchful waiting’ for repairable MR and also conventional versus minimally invasive surgery to assess cost-effectiveness, duration of hospitalisation and clinical outcome. Registries of residual regurgitation after mitral repair are required to establish long-term durability.

Although non-randomised studies in AS suggest that early surgery is superior to watchful waiting in critical disease (transaortic velocity >5.5 m/s),10 ,11 uncertainty remains about the optimal timing of surgery in asymptomatic patients with subcritical disease (transaortic velocity 4.5–5.5 m/s) or elderly patients with critical disease. Novel imaging markers in patients with asymptomatic severe AS have been proposed following observational studies, such as haemodynamic changes during exercise stress echocardiography12 or myocardial fibrosis on MRI,13 but none have been tested in an RCT to determine the clinical utility or impact upon survival.

Secondary or functional MR remains a clinical challenge. Guidelines currently suggest surgery should only be considered after failure of optimal medical therapy (OMT), including resynchronisation therapy.3 ,14 However, mitral annuloplasty, even in patients with very low ejection fraction, may induce left ventricular (LV) reverse remodelling.15 There is a need for an RCT of surgery versus OMT in patients with and without LV remodelling in symptomatic functional MR. Finally, few data on tricuspid valve surgery exist and a large registry would be useful as well as an RCT of tricuspid annuloplasty, guided by annular diameter on echocardiography, performed at the time of left-sided valve surgery.

Infective endocarditis

The absence of an RCT supporting routine antibiotic prophylaxis prior to dental work was one of the key reasons that the UK body the National Institute for Health and Clinical Excellence recommended avoidance of virtually all prophylaxis.16 This is contrary to other international guidelines, which makes the UK an ideal test bed for an RCT. This would be most practical in high-risk patients (prior endocarditis or prosthetic heart valves). In the interim, such patients should be included in case–control studies to document the temporal relation between endocarditis and dental procedures.

The optimal timing of surgery in patients with endocarditis remains unclear, with a relatively high incidence of surgery following hospital discharge. It is not established how vegetation length and mobility should be used to inform surgery. A recent trial in patients with large vegetations and severe valve disease favoured earlier surgery over conventionally timed surgery,17 but similar RCTs are still required in other patient groups.

Prosthetic heart valves

Prosthetic valves are a palliative treatment for VHD, which is best suited to the developed world. Biological valves fail quickly in the young, but mechanical valves currently cannot be implanted without anticoagulation. However, newer bileaflet prosthesis in the aortic position potentially may be manageable with low-level anticoagulation. Larger trials investigating this issue, as well as the role of novel oral anticoagulants, are needed.

Few advances in prosthetic valve technology have been made since the introduction of the bileaflet mechanical valve in 1977. Research is needed to develop novel designs (eg, biopolymers, xenografts seeded with the recipient's own cells and stem-cell-derived grafts). Registries alerting physicians to unexpected early failures are required. Routine pathology facilities see only small numbers of explants and understandably lack the level of specialist expertise needed in order to determine and categorise the failure modes. The establishment of an independent test laboratory for analysis of all prosthesis failures would provide valuable data on the rates and mechanisms of failure. Ideally, controlled trials of all new heart valves against an industry standard should be set up at designated VHD centres to allow standardised collection of clinical, haemodynamic and safety data.

Emerging technologies

Although frequently used in research as an important component of the design process of prosthetic valves, computer modelling techniques are only now sufficiently mature to be applied with confidence in the clinical arena. Improvements in diagnostic imaging are driving the development of patient-specific models, firmly founded on the underpinning physical and engineering principles and enable details of the local flow field and the pressure distribution across the valve to be simulated. For the diseased native valve, prediction of the patient-specific haemodynamic response under stress could provide support for  diagnosis and planning treatment, providing valuable insight into the optimal time for replacement in a given patient. Trialling of these emerging technologies will require new approaches to technology assessment and present new challenges for clinicians.

Organisational

The management of rheumatic VHD in industrially underdeveloped regions requires political and social changes to improve housing and sanitation and greatly increase access to medical care. The Drakensburg declaration18 has already highlighted the need to raise awareness, improve information through epidemiological surveillance and establish national primary and secondary prevention programmes.

Even in industrialised countries, the economic burden of VHD is unknown, but is essential for planning services. Wide variation in access to appropriate VHD specialist care persists19 and the reasons need to be explored. Methods for improving detection of VHD should also be assessed. Specialist valve clinics linked to expert surgical centres with seamless links to the community are now seen as best practice,20 but the supportive evidence is predominantly from the USA.